Trustworthy Multi-Agent Systems: Mitigating Semantic Drift with the Argent Signaling Protocol

Researchers introduced the Argent Signaling Protocol (ASP) to mitigate semantic drift in multi-agent LLM systems by attaching structured quality signals to AI responses. In tests, ASP improved pass rates on a document-grounded QA benchmark from 11.1% to 33.3% on Qwen and blocked 100% of ungrounded outputs in multi-agent mode. The protocol enables controllers to distinguish repairable failures from containment failures, improving system reliability.

arXiv:2606.19356v1 Announce Type: new Abstract: When multi-agent LLM systems produce bad answers, not all failures are equal: some answers are grounded in the right material but incomplete, while others are simply ungrounded and should be stopped. Current retry strategies treat both cases identically try again and hope for the best , leaving human supervisors unable to tell whether a retry was warranted or whether the system should have halted instead. We introduce the Argent Signaling Protocol ASP , a compact machine-readable header that accompanies every AI-generated response with structured quality signals: certainty @C , grounding @G , stochasticity @S , and an assumption index that classifies the evidentiary basis of each claim. These signals enable a controller to distinguish repairable failures from containment failures and route each case differently. We evaluate ASP in two modes. In standalone mode, a 27-question document-grounded QA benchmark over the Array BioPharma/Ono license agreement compares baseline prompts against ASP-instrumented controller actions across three local GGUF models. On Qwen~ 0.8B , ASP improves pass rate from 11.1% to 33.3% and mean term coverage from 36.7% to 65.4%; on Dobby~ 8B , ASP produces 4 fail-to-pass recoveries, raising pass rate from 33.3% to 44.4%; on SmolLM3~ 3B , ASP alternates between repair and containment per question. Aggregate improvement is meaningful 12/81 to 21/81 passes . In multi-agent mode, an ASP sidecar sits between a retrieval agent and a downstream decision agent; the sidecar blocks 100% of ungrounded upstream outputs from reaching the downstream agent 24/27 blocked, 0 ungrounded propagations .